-
A Vodolazkaia,
Y El-Aalamat,
D Popovic,
A Mihalyi,
X Bossuyt,
C M Kyama,
A Fassbender,
A Bokor,
D Schols,
D Huskens,
C Meuleman,
K Peeraer,
C Tomassetti,
O Gevaert,
E Waelkens,
A Kasran, B De Moor,
T M D'Hooghe
[show abstract]
[hide abstract]
ABSTRACT: At present, the only way to conclusively diagnose endometriosis is laparoscopic inspection, preferably with histological confirmation. This contributes to the delay in the diagnosis of endometriosis which is 6-11 years. So far non-invasive diagnostic approaches such as ultrasound (US), MRI or blood tests do not have sufficient diagnostic power. Our aim was to develop and validate a non-invasive diagnostic test with a high sensitivity (80% or more) for symptomatic endometriosis patients, without US evidence of endometriosis, since this is the group most in need of a non-invasive test.
A total of 28 inflammatory and non-inflammatory plasma biomarkers were measured in 353 EDTA plasma samples collected at surgery from 121 controls without endometriosis at laparoscopy and from 232 women with endometriosis (minimal-mild n = 148; moderate-severe n = 84), including 175 women without preoperative US evidence of endometriosis. Surgery was done during menstrual (n = 83), follicular (n = 135) and luteal (n = 135) phases of the menstrual cycle. For analysis, the data were randomly divided into an independent training (n = 235) and a test (n = 118) data set. Statistical analysis was done using univariate and multivariate (logistic regression and least squares support vector machines (LS-SVM) approaches in training- and test data set separately to validate our findings.
In the training set, two models of four biomarkers (Model 1: annexin V, VEGF, CA-125 and glycodelin; Model 2: annexin V, VEGF, CA-125 and sICAM-1) analysed in plasma, obtained during the menstrual phase, could predict US-negative endometriosis with a high sensitivity (81-90%) and an acceptable specificity (68-81%). The same two models predicted US-negative endometriosis in the independent validation test set with a high sensitivity (82%) and an acceptable specificity (63-75%).
In plasma samples obtained during menstruation, multivariate analysis of four biomarkers (annexin V, VEGF, CA-125 and sICAM-1/or glycodelin) enabled the diagnosis of endometriosis undetectable by US with a sensitivity of 81-90% and a specificity of 63-81% in independent training- and test data set. The next step is to apply these models for preoperative prediction of endometriosis in an independent set of patients with infertility and/or pain without US evidence of endometriosis, scheduled for laparoscopy.
Human Reproduction 06/2012; 27(9):2698-711. · 4.47 Impact Factor
-
A Fassbender,
N Verbeeck,
D Börnigen,
C M Kyama,
A Bokor,
A Vodolazkaia,
K Peeraer,
C Tomassetti,
C Meuleman,
O Gevaert,
R Van de Plas,
F Ojeda, B De Moor,
Y Moreau,
E Waelkens,
T M D'Hooghe
[show abstract]
[hide abstract]
ABSTRACT: An early semi-invasive diagnosis of endometriosis has the potential to allow early treatment and minimize disease progression but no such test is available at present. Our aim was to perform a combined mRNA microarray and proteomic analysis on the same eutopic endometrium sample obtained from patients with and without endometriosis.
mRNA and protein fractions were extracted from 49 endometrial biopsies obtained from women with laparoscopically proven presence (n= 31) or absence (n= 18) of endometriosis during the early luteal (n= 27) or menstrual phase (n= 22) and analyzed using microarray and proteomic surface enhanced laser desorption ionization-time of flight mass spectrometry, respectively. Proteomic data were analyzed using a least squares-support vector machines (LS-SVM) model built on 70% (training set) and 30% of the samples (test set).
mRNA analysis of eutopic endometrium did not show any differentially expressed genes in women with endometriosis when compared with controls, regardless of endometriosis stage or cycle phase. mRNA was differentially expressed (P< 0.05) in women with (925 genes) and without endometriosis (1087 genes) during the menstrual phase when compared with the early luteal phase. Proteomic analysis based on five peptide peaks [2072 mass/charge (m/z); 2973 m/z; 3623 m/z; 3680 m/z and 21133 m/z] using an LS-SVM model applied on the luteal phase endometrium training set allowed the diagnosis of endometriosis (sensitivity, 91; 95% confidence interval (CI): 74-98; specificity, 80; 95% CI: 66-97 and positive predictive value, 87.9%; negative predictive value, 84.8%) in the test set.
mRNA expression of eutopic endometrium was comparable in women with and without endometriosis but different in menstrual endometrium when compared with luteal endometrium in women with endometriosis. Proteomic analysis of luteal phase endometrium allowed the diagnosis of endometriosis with high sensitivity and specificity in training and test sets. A potential limitation of our study is the fact that our control group included women with a normal pelvis as well as women with concurrent pelvic disease (e.g. fibroids, benign ovarian cysts, hydrosalpinges), which may have contributed to the comparable mRNA expression profile in the eutopic endometrium of women with endometriosis and controls.
Human Reproduction 05/2012; 27(7):2020-9. · 4.47 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Bias-corrected approximate 100(1-α)% pointwise and simultaneous confidence and prediction intervals for least squares support vector machines are proposed. A simple way of determining the bias without estimating higher order derivatives is formulated. A variance estimator is developed that works well in the homoscedastic and heteroscedastic case. In order to produce simultaneous confidence intervals, a simple Šidák correction and a more involved correction (based on upcrossing theory) are used. The obtained confidence intervals are compared to a state-of-the-art bootstrap-based method. Simulations show that the proposed method obtains similar intervals compared to the bootstrap at a lower computational cost.
IEEE Transactions on Neural Networks 02/2011; · 2.95 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Recently, a Risk of Ovarian Malignancy Algorithm (ROMA) utilising human epididymis secretory protein 4 (HE4) and CA125 successfully classified patients as presenting a high or low risk for epithelial ovarian cancer (EOC). We validated this algorithm in an independent prospective study.
Women with a pelvic mass, who were scheduled to have surgery, were enrolled in a prospective study. Preoperative serum levels of HE4 and CA125 were measured in 389 patients. The performance of each of the markers, as well as that of ROMA, was analysed.
When all malignant tumours were included, ROMA (receiver operator characteristic (ROC)-area under curve (AUC)=0.898) and HE4 (ROC-AUC)=0.857) did not perform significantly better than CA125 alone (ROC-AUC=0.877). Using a cutoff for ROMA of 12.5% for pre-menopausal patients, the test had a sensitivity of 67.5% and a specificity of 87.9%. With a cutoff of 14.4% for post-menopausal patients, the test had a sensitivity of 90.8% and a specificity of 66.3%. For EOC vs benign disease, the ROC-AUC of ROMA increased to 0.913 and for invasive EOC vs benign disease to 0.957.
This independent validation study demonstrated similar performance indices to those recently published. However, in this study, HE4 and ROMA did not increase the detection of malignant disease compared with CA125 alone. Although the initial reports were promising, measurement of HE4 serum levels does not contribute to the diagnosis of ovarian cancer.
British Journal of Cancer 02/2011; 104(5):863-70. · 5.04 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To investigate changes in abdominal circumference (AC) and umbilical artery pulsatility index (UA-PI) with gestation in fetuses with isolated gastroschisis, and to determine whether a relationship exists between UA-PI and fetal AC.
Data from 58 pregnancies with isolated gastroschisis diagnosed at between 24 and 36 weeks' gestation were included in the study. Z-scores were calculated with respect to expected UA-PI values in normal pregnancies after log-transformation. AC-Z-scores were calculated with respect to expected size in normal pregnancies according to a standard chart. Functional linear discriminant analysis (FLDA) was applied to generate 50(th), 5(th) and 95(th) percentile curves for changes in both AC and UA-PI with gestational age in fetuses with gastroschisis. These curves were compared with the standard curves, as were the means. UA-PI was also plotted against AC. For this relationship, a robust Spearman correlation coefficient was obtained with FLDA.
In fetuses with gastroschisis, there was a highly significant negative correlation between UA-PI and AC, normalized for gestation using Z-scores (median correlation coefficient, - 0.289; median P = 0.000023). Moreover, compared with standard curves AC was lower and UA-PI higher in the gestational-age range studied. Both the AC and UA-PI curves showed a significantly different rate of change with gestation compared with the normal ranges. The mean values for fetuses with gastroschisis compared with the standard AC and UA-PI range curves were significantly different for AC throughout gestation, and for UA-PI from 32 weeks' gestation.
In fetal gastroschisis, it is well known that AC tends to be smaller, though UA-PI has not been reported to be abnormal in any consistent way. There is a clear relationship between the fetus's AC for gestation and UA-PI, which is not the case for normally grown fetuses. The data suggest that the growth restriction seen in gastroschisis may be explained by hypoxia, and not simply by the classical explanation of extra-abdominal displacement of the abdominal viscera.
Ultrasound in Obstetrics and Gynecology 01/2011; 38(5):538-42. · 3.01 Impact Factor
-
A Daemen,
L Valentin,
R Fruscio,
C Van Holsbeke,
G B Melis,
S Guerriero,
A Czekierdowski,
D Jurkovic,
W Ombelet,
A Rossi,
I Vergote,
T Bourne, B De Moor,
D Timmerman
[show abstract]
[hide abstract]
ABSTRACT: The aim of this study was to establish when a second-stage diagnostic test may be of value in cases where a primary diagnostic test has given an uncertain diagnosis of the benign or malignant nature of an adnexal mass.
The diagnostic performance with regard to discrimination between benign and malignant adnexal masses for mathematical models including ultrasound variables and for subjective evaluation of ultrasound findings by an experienced ultrasound examiner was expressed as area under the receiver-operating characteristics curve (AUC), sensitivity and specificity. These were calculated for the total study population of 1938 patients with an adnexal mass as well as for subpopulations defined by the certainty with which the diagnosis of benignity or malignancy was made. The effect of applying a second-stage test to the tumors where risk estimation was uncertain was determined.
The best mathematical model (LR1) had an AUC of 0.95, sensitivity of 92% and specificity of 84% when applied to all tumors. When model LR1 was applied to the 10% of tumors in which the calculated risk fell closest to the risk cut-off of the model, the AUC was 0.59, sensitivity 90% and specificity 21%. A strategy where subjective evaluation was used to classify these 10% of tumors for which LR1 performed poorly and where LR1 was used in the other 90% of tumors resulted in a sensitivity of 91% and specificity of 90%. Applying subjective evaluation to all tumors yielded an AUC of 0.95, sensitivity of 90% and specificity of 93%. Sensitivity was 81% and specificity 47% for those patients where the ultrasound examiner was uncertain about the diagnosis (n = 115; 5.9%). No mathematical model performed better than did subjective evaluation among the 115 tumors where the ultrasound examiner was uncertain.
When model LR1 is used as a primary test for discriminating between benign and malignant adnexal masses, the use of subjective evaluation of ultrasound findings by an experienced examiner as a second-stage test in the 10% of cases for which the model yields a risk of malignancy closest to its risk cut-off will improve specificity without substantially decreasing sensitivity. However, none of the models tested proved suitable as a second-stage test in tumors where subjective evaluation yielded an uncertain result.
Ultrasound in Obstetrics and Gynecology 01/2011; 37(1):100-6. · 3.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To evaluate if the addition of lidocaine to the gel used for gel infusion sonohysterography (GIS) reduces pain experienced during GIS or subsequent hysteroscopy.
A total of 142 consecutive patients were randomized using computer-generated random integers. In 79 patients, GIS was performed with a gel containing lidocaine (Instillagel®) and in 63 patients the gel did not contain lidocaine (Endosgel®). Immediately after GIS, 132 patients (94%) underwent office hysteroscopy. The women were asked to fill in a questionnaire including a 100-mm visual analogue scale (VAS) score after each examination.
The mean age (SD) was 50.8 (12.1) years; 58.5% were premenopausal and 15.6% were nulliparous. The median (interquartile range (IR)) VAS score during GIS for all women was 6 (19.5): 8 (21) for the lidocaine group versus 5 (18.2) for those who received gel without lidocaine. The median (IR) VAS scores during hysteroscopy in the total group, the Instillagel group and the Endosgel group were 15.5 (43.2), 24 (35) and 9 (52), respectively. None of the differences were statistically significant.
The addition of lidocaine to the gel used either for GIS or prior to office hysteroscopy does not reduce the procedure-related pain.
Gynecologic and Obstetric Investigation 01/2011; 71(4):236-9. · 1.28 Impact Factor
-
A Smeets,
A Daemen,
I Vanden Bempt,
O Gevaert,
B Claes,
H Wildiers,
R Drijkoningen,
P Van Hummelen,
D Lambrechts, B De Moor,
P Neven,
C Sotiriou,
T Vandorpe,
R Paridaens,
M R Christiaens
[show abstract]
[hide abstract]
ABSTRACT: The aim of this study was to investigate whether lymph node involvement in breast cancer is influenced by gene or miRNA expression of the primary tumor. For this purpose, we selected a very homogeneous patient population to minimize heterogeneity in other tumor and patient characteristics. First, we compared gene expression profiles of primary tumor tissue from a group of 96 breast cancer patients balanced for lymph node involvement using Affymetrix Human U133 Plus 2.0 microarray chip. A model was built by weighted Least-Squares Support Vector Machines and validated on an internal and external dataset. Next, miRNA profiling was performed on a subset of 82 tumors using Human MiRNA-microarray chips (Illumina). Finally, for each miRNA the number of significant inverse correlated targets was determined and compared with 1000 sets of randomly chosen targets. A model based on 241 genes was built (AUC 0.66). The AUC for the internal dataset was 0.646 and 0. 651 for the external datasets. The model includes multiple kinases, apoptosis-related, and zinc ion-binding genes. Integration of the microarray and miRNA data reveals ten miRNAs suppressing lymph node invasion and one miRNA promoting lymph node invasion. Our results provide evidence that measurable differences in gene and miRNA expression exist between node negative and node positive patients and thus that lymph node involvement is not a genetically random process. Moreover, our data suggest a general deregulation of the miRNA machinery that is potentially responsible for lymph node invasion.
Breast Cancer Research and Treatment 11/2010; 129(3):767-76. · 4.43 Impact Factor
-
Ultrasound in Obstetrics and Gynecology 10/2010; 36(S1):137-138. · 3.01 Impact Factor
-
Ultrasound in Obstetrics and Gynecology 10/2010; 36(S1):229. · 3.01 Impact Factor
-
Ultrasound in Obstetrics and Gynecology 10/2010; 36(S1):8. · 3.01 Impact Factor
-
Ultrasound in Obstetrics and Gynecology 10/2010; 36(S1):57-58. · 3.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We sought to define the relationship between first trimester fetal growth, pregnancy-associated plasma protein A (PAPP-A) levels and birthweight.
Two-hundred and one women with repeat first trimester crown-rump length (CRL) measurements were included. In 194, the first trimester PAPP-A value was known and in 169 there was complete data including birthweight. Fetal growth curves were derived using functional linear discriminant analysis (FLDA) and growth compared between those with < 10th percentile, 10th to 90th and > 90th percentile PAPP-A multiple of median (MoM) levels and birthweight percentiles.
Median maternal age was 35 years, gestation at PAPP-A sampling and of first scan was 11 weeks. Median delivery gestation was 40 weeks and birthweight 3425 g. There was no association between first trimester fetal CRL growth and either PAPP-A MoM percentile or birthweight percentile. There was a significant positive correlation between PAPP-A MoM and birthweight percentile (p = 0.0004).
First trimester fetal growth rate is not related to birthweight percentile or first trimester PAPP-A levels. Irrespective of gestation, a low PAPP-A is associated with delivery of a smaller baby, and a high PAPP-A with a larger baby.
Prenatal Diagnosis 09/2010; 30(9):873-8. · 2.11 Impact Factor
-
A Daemen,
D Jurkovic,
C Van Holsbeke,
S Guerriero,
A C Testa,
A Czekierdowski,
R Fruscio,
D Paladini,
P Neven,
A Rossi,
T Bourne, B De Moor,
D Timmerman
[show abstract]
[hide abstract]
ABSTRACT: Two logistic regression models have been developed for the characterization of adnexal masses. The goal of this prospective analysis was to see whether these models perform differently according to the prevalence of malignancy and whether the cut-off levels of risk assessment for malignancy by the models require modification in different centers.
Centers were categorized into those with a prevalence of malignancy below 15%, between 15 and 30% and above 30%. The areas under the receiver-operating characteristics curves (AUC) were compared using bootstrapping. The optimal cut-off level of risk assessment for malignancy was chosen per center, corresponding to the highest sensitivity level possible while still keeping a good specificity.
Both models performed better in centers with a lower prevalence of malignant cases. The AUCs of the two models for centers with fewer than 15% malignant cases were 0.97 and 0.95, those of centers with 15-30% malignancy were 0.95 and 0.93 and those of centers with more than 30% malignant cases were 0.94 and 0.92. This decrease in performance was due mainly to the decrease in specificity from over 90 to around 76%. In the centers with a higher percentage of malignant cases, a sensitivity of at least 90% with a good specificity could not be obtained by choosing a different cut-off level.
Overall the models performed well in all centers. The performance of the logistic regression models worsened with increasing prevalence of malignancy, due to a case mix with more borderline and complex benign masses seen in those centers. Because the cut-off of 0.10 is optimal for all three types of center, it seems reasonable to use this cut-off for both models in all centers.
Ultrasound in Obstetrics and Gynecology 09/2010; 37(2):226-31. · 3.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The Robinson and Hadlock crown-rump length (CRL) curves are commonly used to estimate gestational age (GA) based on the CRL of an embryo or fetus. However, the Robinson curve was derived from a small population using transabdominal sonography and the Hadlock curve was generated using early transvaginal ultrasound equipment. The aim of this study was to use transvaginal and transabdominal ultrasound to study a large population of early pregnancies to assess embryonic or fetal size, and so create a new normal CRL curve from 5.5 weeks' gestation. We compared this with the Robinson and Hadlock CRL curves.
A retrospective database study of CRL in first-trimester embryos was conducted in a fetal medicine referral center with a predominantly Caucasian population. Linear mixed-effects analysis was performed to determine the relationship between CRL and GA. After internal validation of this curve, the CRL was compared with the expected CRL at a given GA according to both the Robinson and Hadlock models based on the paired t-test. Bland-Altman plots were constructed to compare the CRL measurements obtained in our study population with those predicted according to GA by both the Robinson and Hadlock curves.
In total 3710 normal singleton pregnancies with a known last menstrual period were included in the study, corresponding to 4387 scans. Our data differed significantly from both the Robinson and the Hadlock curves (paired t-test, P < 0.0001). A mixed-effects model for CRL as a function of GA was developed on 70% of the data and internally validated with z-scores on the remaining 30%. The new curve extended from 5.5 to 14 weeks' gestation. Compared to our CRL curve, the Robinson curve gave a 4-day underestimation of GA at 6 weeks with a difference in CRL of 3.7 mm and a 1-day overestimation from 11 to 14 weeks with a difference in CRL of 0.9-1 mm. A comparison between our curve and the Hadlock curve showed a difference in CRL of 2.7 mm at 6 weeks, equivalent to an underestimation of 3 days, and a difference in CRL of 4.8 mm at 14 weeks, equivalent to an overestimation of 2 days. At 9 weeks all three curves were similar.
The new CRL curve suggests differences in the range of CRL measurements compared with the Robinson and Hadlock curves. These differences are most significant at the beginning and the end of the first trimester, and may lead to more accurate estimations of GA.
Ultrasound in Obstetrics and Gynecology 06/2010; 35(6):650-5. · 3.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: In kernel based regression techniques (such as Support Vector Machines or Least Squares Support Vector Machines) it is hard to analyze the influence of perturbed inputs on the estimates. We show that for a nonlinear black box model a convex problem can be derived if it is linearized with respect to the influence of input perturbations. For this model an explicit prediction equation can be found. The cast into a convex problem is possible as we assume that the perturbations are bounded by a design parameter ¿. The problem requires the solution of linear systems in Nd (the number of training points times the input dimensionality) variables. However, approximate solutions can be obtained with moderate computational effort. We demonstrate on simple examples that possible applications are in robust model selection, experiment design and model analysis.
Decision and Control, 2009 held jointly with the 2009 28th Chinese Control Conference. CDC/CCC 2009. Proceedings of the 48th IEEE Conference on; 01/2010
-
A Mihalyi,
O Gevaert,
C M Kyama,
P Simsa,
N Pochet,
F De Smet, B De Moor,
C Meuleman,
J Billen,
N Blanckaert,
A Vodolazkaia,
V Fulop,
T M D'Hooghe
[show abstract]
[hide abstract]
ABSTRACT: Lack of a non-invasive diagnostic test contributes to the long delay between onset of symptoms and diagnosis of endometriosis. The aim of this study was to evaluate the combined performance of six potential plasma biomarkers in the diagnosis of endometriosis.
This case-control study was conducted in 294 infertile women, consisting of 93 women with a normal pelvis and 201 women with endometriosis. We measured plasma concentrations of interleukin (IL)-6, IL-8, tumour necrosis factor-alpha, high-sensitivity C-reactive protein (hsCRP), and cancer antigens CA-125 and CA-19-9. Analyses were done using the Kruskal-Wallis test, Mann-Whitney test, receiver operator characteristic, stepwise logistic regression and least squares support vector machines (LSSVM).
Plasma levels of IL-6, IL-8 and CA-125 were increased in all women with endometriosis and in those with minimal-mild endometriosis, compared with controls. In women with moderate-severe endometriosis, plasma levels of IL-6, IL-8 and CA-125, but also of hsCRP, were significantly higher than in controls. Using stepwise logistic regression, moderate-severe endometriosis was diagnosed with a sensitivity of 100% (specificity 84%) and minimal-mild endometriosis was detected with a sensitivity of 87% (specificity 71%) during the secretory phase. Using LSSVM analysis, minimal-mild endometriosis was diagnosed with a sensitivity of 94% (specificity 61%) during the secretory phase and with a sensitivity of 92% (specificity 63%) during the menstrual phase.
Advanced statistical analysis of a panel of six selected plasma biomarkers on samples obtained during the secretory phase or during menstruation allows the diagnosis of both minimal-mild and moderate-severe endometriosis with high sensitivity and clinically acceptable specificity.
Human Reproduction 12/2009; 25(3):654-64. · 4.47 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To compare gel infusion sonohysterography (GIS) with saline contrast sonohysterography (SCSH) with regard to technical feasibility and procedure-related pain experienced by patients.
This prospective observational cohort study included 551 consecutive patients with abnormal bleeding: SCSH was attempted in the first 402 women and GIS was attempted in the following 149. All procedures were performed by the same examiner, in the same clinical setting, using a 2-mm diameter catheter. After contrast sonohysterography, most patients underwent office hysteroscopy (n = 502) and endometrial sampling (n = 323). The women were asked to rate the pain experienced during each procedure using a 100-mm visual analog scale (VAS). Patients' characteristics, ultrasound findings, histological diagnosis, technical failures and procedure-related pain were compared between the two procedures.
The mean +/- SD VAS score for contrast sonography, subsequent hysteroscopy and endometrial biopsy were 22.9 +/- 21.7, 38.8 +/- 26.6 and 50.0 +/- 26.3, respectively, in the SCSH subgroup vs. 16.5 +/- 21.5, 27.6 +/- 28 and 33.6 +/- 30.3, respectively, in the GIS subgroup (P = 0.0051, P = 0.0005 and P = 0.0003, respectively). The technical failure rate was 5% in the SCSH subgroup vs. 2% in the GIS subgroup (P = 0.1522).
GIS and SCSH showed similar technical feasibility. The procedure-related pain reported by patients during contrast sonohysterography, as well as during subsequent hysteroscopy and endometrial sampling, was less in the GIS group.
Ultrasound in Obstetrics and Gynecology 11/2009; 34(6):711-4. · 3.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: For most diseases and examinations, clinical data such as age, gender and medical history guides clinical management, despite the rise of high-throughput technologies. To fully exploit such clinical information, appropriate modeling of relevant parameters is required. As the widely used linear kernel function has several disadvantages when applied to clinical data, we propose a new kernel function specifically developed for this data. This ldquoclinical kernel functionrdquo more accurately represents similarities between patients. Evidently, three data sets were studied and significantly better performances were obtained with a Least Squares Support Vector Machine when based on the clinical kernel function compared to the linear kernel function.
Engineering in Medicine and Biology Society, 2009. EMBC 2009. Annual International Conference of the IEEE; 10/2009
-
Ultrasound in Obstetrics and Gynecology 10/2009; 34(S1):85. · 3.01 Impact Factor
